Research Article

Enhancement of Executive Functions but Not Memory by Multidomain Group Cognitive Training in Patients with Parkinson’s Disease and Mild Cognitive Impairment: A Multicenter Randomized Controlled Trial

Table 4

Regression analyses: predictors for intervention success.

ModelAnalysisF-testR2 (adjusted R2)Standardized coefficients (β) of significant predictors
Baseline levelAge at baselineEducation in yearsUPDRS-3 at baselineLEDD at baselineApoE4 carrier

Cognitive training
Memory
 ROCFT delayed recallaITTF (6, 25) = 2.688; 0.392 (0.246)−0.506
PPF (6, 22) = 3.618; 0.497 (0.359)−0.583−0.373†↓−0.335†↓
Executive functions
 Executive functions mean scoreaITTF (6, 22) = 2.673; 0.422 (0.264)−0.446
PPn.s.
 Semantic word fluencyaITTF (6, 25) = 2.643; 0.388 (0.241)−0.3980.493
PPF (6, 23) = 2.865; 0.428 (0.278)−0.4780.546
 Phonemic word fluencyaITTF (6, 25) = 3.595; 0.463 (0.334)−0.5870.405†↑−0.5070.4230.473
PPF (6, 22) = 3.660; 0.500 (0.363)−0.592−0.340†↓−0.4400.4720.500
 MCST categoriesaITTF (6, 22) = 2.797; 0.433 (0.278)−0.562−0.374†↓
PPF (6, 20) = 2.837; 0.460 (0.298)−0.580
 Key searchaITTF (6, 25) = 2.778; 0.400 (0.256)−0.660
PPF (6, 23) = 2.651; 0.409 (0.255)−0.660
Working memory
 Working memory mean scoreaITTF (6, 25) = 2.705; 0.394 (0.248)−0.622
PPF (6, 23) = 2.670; 0.411 (0.257)−0.633
 Digit span backward (WAIS)aITTF (6, 25) = 5.311; 0.560 (0.455)−0.729
PPF (6, 23) = 5.700; 0.598 (0.493)−0.744
Visuocognition
 Visuocognition mean scoreaITTF (6, 25) = 5.052; 0.548 (0.440)−0.5480.334†↑
PPF (6, 23) = 4.770; 0.554 (0.438)−0.552
 ROCFT figure copyaITTF (6, 25) = 3.460; 0.454 (0.323)−0.515
PPF (6, 23) = 3.296; 0.462 (0.322)−0.537
 Benton judgment of line orientationaITTF (6, 25) = 5.049; 0.548 (0.439)−0.722
PPF (6, 23) = 5.009; 0.567 (0.453)−0.726
Language
 Language mean scoreaITTF (6, 25) = 3.012; 0.420 (0.280)−0.565−0.4410.463
PPF (6, 23) = 2.786; 0.421 (0.270)−0.555−0.3920.407†↑
 Speech comprehension (ACL)aITTF (6, 25) = 3.842; 0.480 (0.355)−0.642
PPF (6, 23) = 3.672; 0.489 (0.356)−0.655
Noncognitive outcomes
 BDI-IIbITTF (6, 25) = 2.858; 0.407 (0.264)−0.473−0.365†↑
PPF (6, 23) = 2.724; 0.415 (0.263)−0.489
 UPDRS-IIIbITTF (5, 26) = 3.281; 0.387 (0.269)−0.328†↑0.403
PPF (5, 24) = 4.583; 0.488 (0.382)−0.3720.467−0.376†↑

Control group
Memory
 CVLT total score trials 1–5aITTF (6, 20) = 2.978; 0.472 (0.313)−0.702−0.456†↓
PPF (6, 19) = 3.051; 0.491 (0.330)−0.690−0.429†↓
 CVLT long delay free recall IIaITTF (6, 20) = 3.219; 0.491 (0.339)−0.362−0.776−0.637
PPF (6, 19) = 3.042; 0.490 (0.329)−0.346†↓−0.777−0.599
 ROCFT delayed recallaITTF (6, 21) = 4.872; 0.582 (0.462)−0.5270.462
PPF 6, 20) = 4.640; 0.582 (0.456)−0.5230.459
Executive functions
 Key searchaITTF (6, 21) = 4.075; 0.538 (0.406)−0.780
PPF (6, 20) = 3.942; 0.542 (0.404)−0.785
Visuocognition
 Visuocognition mean scoreaITTF (6, 21) = 2.972; 0.459 (0.305)−0.654
PPF (6, 20) = 3.285; 0.496 (0.345)−0.667
 ROCFT figure copyaITTF (6, 21) = 5.361; 0.605 (0.492)−0.796
PPF (6, 20) = 5.638; 0.628 (0.512)−0.802
Language
 Language mean scoreaITTF (6, 21) = 9.890; 0.739 (0.664)−0.847
PPF (6, 19) = 9.417; 0.748 (0.669)−0.856
 Boston naming testaITTF (6, 21) = 6.582; 0.653 (0.554)−0.678
PPF (6, 20) = 6.563; 0.663 (0.562)−0.619
 Speech comprehension (ACL)aITTF (6, 21) = 25.963; 0.881 (0.847)−0.900−0.196
PPF (6, 19) = 23.387; 0.881 (0.843)−0.899−0.187†↓

Dependent variables are defined as Δposttest − pretest; only those regression models that reached statistical significance at are presented; for each significant regression model, standardized regression coefficients are reported for predictors that have reached statistical significance only; ; ; ; . ACL, Aphasia Check List; ApoE, apolipoprotein E; BDI, Beck Depression Inventory; CVLT, California Verbal Learning Test; ITT, intention-to-treat analysis; LEDD, levodopa equivalent daily dose; MCST, Modified Wisconsin Card Sorting Test; PP, per-protocol analysis; ROCFT, Rey–Osterrieth Complex Figure Test; UPDRS, Unified Parkinson’s Disease Rating Scale; WAIS, Wechsler Adult Intelligence Scale. aHigher change scores indicate higher training gains; blower change scores indicate higher training gains. Lower predictor level predicts higher training gains. Higher predictor level predicts higher training gains.